In existing computed tomography (CT) reconstruction images, positions of area to be reconstructed are reconstructed using sampling data of an image acquired by a sensor, and thereby interface information about oversampling, etc., limited to a size of one pixel that is unique to the sensor, is minimally expressed.
In image systems using X-rays, an image is blurred due to several principles that are necessary in the process of acquiring the image, and a reconstruction image using the blurred image is also limited to its resolution.
A blurring model of this image system is expressed by Equation 1. A subject of an original shape passes through the image system, and the image system outputs a final result influenced by its characteristic function, noise, and so on.g(x,y)=h(x,y)*f(x,y)  Equation 1
Here, f(x,y) indicates an amount of light of an original image at x and y-axial coordinates (x, y), h(x,y) is a point spread function as a blurring function, and g(x,y) indicates an amount of light of a blurred original image.
This blurring function produces an effect of blurring an interface of the subject as in FIG. 2 due to a penumbral effect caused by a size of a focal point, and has an effect of fundamentally reducing and limiting a resolution of the sensor and a resolution of the image ahead of the other limitations.